The Standardized Infection Ratio Linda R Greene, RN, MPS,CIC Rochester General Health System...
-
Upload
oliver-todd-norris -
Category
Documents
-
view
216 -
download
1
Transcript of The Standardized Infection Ratio Linda R Greene, RN, MPS,CIC Rochester General Health System...
The Standardized Infection Ratio
Linda R Greene, RN, MPS,CICRochester General Health System
Rochester, [email protected]
Objectives
• Describe what the Standardized Infection Ratio (SIR) is and how it is calculated.
• Explain how to generate and interpret a report
utilizing the SIR. • Identify uses for the SIR in public reporting • Explain the relationship between HAI rates and
the SIR
• Standardized Infection Ratio ( SIR) is a summary measure used to compare the HAI experience among one or more groups of patients to that of a standard population’s (e.g. NHSN)
• Indirect standardization method- Comparison to a referent population
Standardized Infection Ratio Method
What is a standardized infection ratio (SIR)?
• The standardized infection ratio (SIR) is a summary measure used to track HAIs at a national, state, or local level over time
• The SIR adjusts for patients of varying risk within each facility
• It is a summary statistic widely used in public health• In HAI data analysis, the SIR compares the actual
number of HAIs reported with the baseline U.S. experience
I was just getting used to rates, why the SIR?
More sensitive for low denominators
Ability to combine data
Useful for predicting state and national rates
OK , I’m no statistician what’s all this mumbo jumbo about?
In simple terms- you are compared to the average of a referent population adjusted for risk. In this case it is a historical control.
The SIR
Let’s take a closer look
Hospital A :
Type of ICU Number of Infections
Line days My rate NHSN Mean
Med/ Surg 1 865 1.1 2.1
SICU 0 1000 0 2.8
CTICU 2 1065 1.8 1.1
MICU 2 1000 2.0 2.1
Turned into SIR
How do we get the expected ?Type of ICU Number of
InfectionsLine days My rate NHSN Mean
Med/ Surg 1 865 1.1 2.1
SICU 0 1000 0 2.8
CTICU 2 948 2.1 1.1
MICU 2 1000 2.0 2.1
Med Surg 2.1 /1000 x 865= 0.95
SICU 2.8 /1000 X 1000= 2.8CTICU 1.1/1000 X 848= 0.93 MICU 2.1 / 1000 X1000= 2.1
The SIR
Type of ICU Number of infections
Number expected
SIRObserved/expected
P VALUE
Med/ Surg 1 0.95 1.05
SICU 0 2.8 0
CTICU 2 0.93 2.1
MICU 2 2.1 0.95
5 6.78 0.7
SIR is less than 1
Simply Put Simply Put
• A SIR of 1.0 means the observed number of infections is equal to the number of expected infections.
• A SIR above 1.0 means that the infection rate is higher than that found in the "standard population." For HAI reports, the standard population comes from data reported by the hundreds of U.S. hospitals that use the NHSN system. The difference above 1.0 is the percentage by which the infection rate exceeds that of the standard population.
• A SIR below 1.0 means the infection rate is lower than that of the standard population. The difference below 1.0 is the percentage by which the infection rate is lower than that experienced by the standard population
Statistical Significance
• If the P value is less than .05 then your rates are different than the national average
• If the confidence level does not overlap 1, then your rates are different than the national
average.
States with Mandatory HAI Laws
Conducting your own analysis
orgid=10330
Surgical SIR
Calculation
Observed ( number of Infections) Expected (expected number of infections)
Surgery data vs. CLABSI
•Uses patient level data •Logistic regression modeling•Excludes superficial infections
Example
SSI SIR
Interpreting the SIR
The SIRPROS CONS
Surgical risk adjustment is a significant improvement
Risk adjustment still suboptimal – especially with CLABSI data
Consistent with other types of data such as mortality
Not designed to compare 1 institution to another- only to compare with national average
Advantages with rare events Potential problems with ranking ,etc
Overall rates can cloud the big picture
Colon SSI per Month 2010- 2011
Using data Locally
2010 SSI Expected and Observed SSI
NumberOfInfections
STILL FINALIZING DATA – MORE ANALYSIS TO GO
Questions ?